Clinical trials of topical agents: Easy application but difficult design

Superficial thrombophlebitis is an inflammatory disorder of superficial veins and venous thrombosis. Its prevalence in lower limbs is 60%–80% and in the small/short saphenous vein is 10%–20%. It can lead to venous thromboembolic disorders, mainly deep-venous thrombosis and pulmonary embolism.[1] Infusion thrombophlebitis, which is due to catheter insertion, venipuncture, or intravenous drug infusion, is mainly seen in hospitalized patients. The duration of catheterization, catheter-related infection, and catheter material is associated with peripheral vein infusion thrombophlebitis.[2] Although there are guidelines for the treatment of thrombophlebitis, these guidelines do not specify how to prevent infusion-associated superficial thrombophlebitis. Bajpai et al. in their article compared the efficacy, safety, and cost-effectiveness of quick penetrating solution (QPS) heparin, QPS diclofenac, and heparin gel in the prevention of superficial thrombophlebitis. They conducted a randomized controlled trial enrolling 219 patients who were 18–60 years age and needed venous cannulation for minimum of 72 h. Patients randomly received the three study drugs every 8 h for a period of 72 h. The Visual Infusion Phlebitis Scale was used to grade the venous cannulation site. The Kaplan–Meier analysis did not show any significant differences for the prevention of any grade ST or severe ST in three treatment arms. The mean cost-effectiveness ratio for preventing thrombophlebitis was 14.2 in heparin gel, 13.2 in QPS heparin, and 95.6 in QPS diclofenac-treated patients.[4] Heparin and diclofenac have been used for a very long time for prevention; however, there are not many studies comparing the efficacy and safety of the above three drugs. A Cochrane review conducted in 2015 to summarize the evidence from randomized clinical trials concerning the efficacy and safety of medical therapy of superficial thrombophlebitis of the upper extremity concluded that the evidence is limited and the data are of low quality. Data were too preliminary to evaluate the effectiveness and safety of topical treatments, systemic anticoagulation, or oral nonsteroidal anti-inflammatory drugs.[3] There has been no such Cochrane review conducted on this topic after that. There were two trials conducted in the past 10 years. One was a randomized, controlled, and double-blind study comparing heparin QPS and diclofenac quick penetrating solution for the prevention of superficial thrombophlebitis caused by IV cannulation. This study showed that heparin QPS is more effective than diclofenac QPS.[5] In another prospective, randomized study which assessed safety and efficacy of heparin topical solution (1000 IU/ml) as compared to heparin topical gel (200 IU/g) in the prevention of superficial thrombophlebitis due to IV infusion concluded that heparin QPS was more efficacious than heparin gel in the prevention of infusion-associated phlebitis. Both had a safety profile.[6] The current study by Bajpai et al. has been well conducted. However, as specified by the authors, blinding was tough since the three study medications were available in different formulations.[4] Blinding is intended to decrease the potential biases which result from differences in management, treatment, or assessment of patients, or interpretation of results that could arise as a result of subject or investigator knowledge of the assigned treatment. In a double-blind clinical trials, participants and investigator and/or investigator’s staff who are the part of the treatment administration or clinical evaluation of subjects are not aware of each participant’s assigned treatment.[7,8] In this study, study team’s nursing staff involved in medication administration and participants were aware of the treatment allocation after the informed consent. Trained nursing staff who were not part of the site team evaluated the patients for the development of thrombophlebitis. Randomization codes and allocation were not revealed to the outcome assessors and data entry team, so they were blinded to the intervention.[4] A few more ways in which the independent assessments could have been done would be to have an independent adjudication committee who would assess the local symptoms of superficial thrombophlebitis using the standardized photos. The study used trained nursing staff for assessing thrombophlebitis which could have been a bias since they were the part of the investigational site. Visual infusion phlebitis score was used for assessing thrombophlebitis. This is a well published and acceptable method of assessment. However, the authors did not specify the version and the source of this score and the validation of this. Cost-effectiveness was analyzed in this study which is novel since this has not been done for these three prevention modalities. However, both direct and indirect costs were not evaluated. It would be good to present this analysis. In this study, only the cost of the three active treatments was compared and cost-effectiveness ratio was calculated.[4] The analysis should also assess nonhealthcare costs, such as loss of work or productivity, informal care, and out-of-pocket expenses.[9] It would be good to find statistical difference in the cost analysis using T-test which was not done in this study. The safety profile of the three formulations has not been evaluated and presented in the study. This would have provided added value to the study. Since there are no recent studies which compared the effectiveness of the three treatments in the prevention of superficial thrombophlebitis, it is good to have a randomized controlled trial assessing the three medications. However, we need further well-designed studies to understand the prevention of infusion-associated superficial thrombophlebitis which also considers the issues related to comparing drugs which are difficult to blind and mask. Further for cost-effectiveness analysis, a deep dive is required into the different aspects of health economics research. DISCLOSURE The insights stated in this article are author’s personal opinion and does not reflect those of the current and previous employers.